Neural networks to generate reliability scores

Abstract

Systems and methods are described relating to score generation for an account of a delivery service using one or more neural networks. The one or more neural networks can generate the score using time series data including metrics associated with the account. In response to a time slot being selected by the account from the set of time slots, the score can be used to generate safeguards for the account, where the safeguards are to be applied prior to execution of one or more scheduled deliveries associated with the time slot.

What is claimed is: 
     
         1 . A computer-implemented method, comprising:
 generating a plurality of time slots that correspond to one or more scheduled deliveries from a station;   causing a client device associated with a carrier of a delivery service to indicate the plurality of time slots;   receiving, from the client device, a selection of a time slot of the plurality of time slots;   determining that the client device is located in a geographical region that includes the station;   obtaining a first set of time series data and a second set of time series data, wherein the first set of time series data and the second set of time series data comprise metrics of the carrier recorded in a plurality of periods;   identifying a first weight for the first set of time series data and a second weight for the second set of time series data;   generating, using one or more neural networks, a score of the carrier based, at least in part, on the first weight, the first set of time series data, the second weight, and the second set of time series data;   identifying one or more safeguards to be applied to the selected time slot based, at least in part, on the score of the carrier, wherein the identification of the one or more safeguards is dependent on the score of the carrier such that a higher score results in one or more fewer safeguards being applied to the carrier, the one or more fewer safeguards corresponding to fewer communications to the carrier to execute a scheduled task corresponding to the selected time slot; and   controlling access to information from the client device to execute the scheduled delivery corresponding to the selected time slot based, at least in part, on compliance with the one or more safeguards.   
     
     
         2 . The computer-implemented method of  claim 1 , further comprising:
 obtaining, from the client device, identification information that comprises driver license information, wherein the identification information is associated with at least one of the one or more safeguards.   
     
     
         3 . The computer-implemented method of  claim 1 , further comprising:
 determining that one or more responses from the client device comply with the one or more safeguards; and   causing the client device to indicate instructions for the carrier to execute the scheduled delivery corresponding to the selected time slot.   
     
     
         4 . The computer-implemented method of  claim 1 ,
 wherein the metrics of the carrier comprise delivery accuracy or check-in time.   
     
     
         5 . A system, comprising:
 one or more processors; and   memory that stores computer-executable instructions that, if executed, cause the one or more processors to:   generate a set of time slots that comprises scheduled deliveries;   cause a client device associated with an account of a service to provide an interface that indicates the set of time slots;   receive, from the client device, a selection of a time slot of the set of time slots;   obtain a first set of time series data and a second set of time series data that comprise a set of metrics of the account recorded in a plurality of periods;   identify one or more weights to be assigned to the first set of time series data and the second set of time series data;   generate, using one or more neural networks, a score associated with the account based, at least in part, on the first set of time series data and the second set of time series data assigned with the one or more weights;   identify one or more safeguards to be applied to the selected time slot prior to executing at least one scheduled delivery in the selected time slot based, at least in part, on the score associated with the account such that a higher score results in one or more fewer safeguards being applied to account, the one or more fewer safeguards corresponding to fewer communications to the client device to execute one or more scheduled tasks in the selected time slot; and   control access to information from the client device based, at least in part, on compliance with the one or more safeguards, wherein the information is to enable one or more operations to execute the at least one scheduled delivery at the selected time slot.   
     
     
         6 . The system of  claim 5 , wherein the computer-executable instructions further comprise computer-executable instructions that, if executed by the one or more processors, cause the system to receive, from the client device, identification data in response to applying the one or more safeguards to the account. 
     
     
         7 . The system of  claim 5 , wherein the computer-executable instructions further comprise computer-executable instructions that, if executed by the one or more processors, cause the system to cause the client device to provide instructions to execute the at least one scheduled delivery in the time slot as a result of one or more verified responses received from the client device. 
     
     
         8 . The system of  claim 5 , wherein the computer-executable instructions further comprise computer-executable instructions that, if executed by the one or more processors, cause the system to modify, using the one or more neural networks, the score based, at least in part, on receiving an indication that the at least one scheduled delivery of the time slot was uncompleted. 
     
     
         9 . The system of  claim 5 , the computer-executable instructions further comprise computer-executable instructions that, if executed by the one or more processors, cause the system to determine that the client device is located in a geofence region that comprises the location. 
     
     
         10 . The system of  claim 5 , wherein the set of metrics comprises delivery accuracy or check-in time. 
     
     
         11 . The system of  claim 5 , wherein the one or more neural networks are updated based, at least in part, on data received from a service that receives feedback from one or more customers associated with at least the set of time slots. 
     
     
         12 . The system of  claim 5 , wherein the one or more neural networks comprise a recurrent neural network (RNN). 
     
     
         13 . A non-transitory computer-readable storage medium storing thereon executable instructions that, as a result of being executed by one or more processors of a computer system, cause the computer system to at least:
 receive, from a service, first information of a set of time slots that corresponds to a set of scheduled deliveries from a location, wherein the first information is associated with an account;   indicate the set of time slots on a screen based, at least in part, on the information;   transmit, to the service, a selection of a time slot of the set of time slots; and   receive second information of a set of protocols to be applied to the account prior to executing one or more scheduled deliveries of the time slot, wherein:
 the set of protocols to be applied to the selected time slot are selected based, at least in part, on a score that is associated with the account such that a higher score results in fewer communications to the account to execute one or more tasks of the one or more scheduled deliveries corresponding to the selected time slot, wherein compliance with the set of protocols causes access to information from the account to execute a scheduled delivery corresponding to the selected time slot to be controlled; 
 the score is generated using one or more neural networks that receive two or more sets of data that are associated with the account; and 
 the two or more sets of data are modified based, at least in part, on a set of weights. 
   
     
     
         14 . The non-transitory computer-readable storage medium of  claim 13 , wherein the data comprises metrics measured in a plurality of periods. 
     
     
         15 . The non-transitory computer-readable storage medium of  claim 13 , wherein the instructions further comprise instructions that, as a result of being executed by the one or more processors, cause the computer system to transmit, to the service, one or more responses generated as a result of complying with the set of protocols. 
     
     
         16 . The non-transitory computer-readable storage medium of  claim 13 , wherein the instructions further comprise instructions that, as a result of being executed by the one or more processors, cause the computer system to display, via a graphical user interface (GUI), instructions to execute one or more scheduled deliveries of the time slot based, at least in part, on receiving an indication that the account complied with the set of protocols. 
     
     
         17 . The non-transitory computer-readable storage medium of  claim 13 , wherein one or more weights are assigned to the two or more sets of time series data. 
     
     
         18 . The non-transitory computer-readable storage medium of  claim 13 , wherein the set of metrics comprise check-in time or delivery accuracy. 
     
     
         19 . The non-transitory computer-readable storage medium of  claim 13 , wherein the instructions further comprise instructions that, as a result of being executed by the one or more processors, cause the computer system to indicate, to the service, that the computer system is within a geofenced region that comprises the location. 
     
     
         20 . The non-transitory computer-readable storage medium of  claim 13 , wherein the one or more neural networks comprise a long short-term memory (LSTM) network.